16:00:38 #startmeeting blazar 16:00:39 Meeting started Thu Mar 11 16:00:38 2021 UTC and is due to finish in 60 minutes. The chair is priteau. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:00:40 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 16:00:42 The meeting name has been set to 'blazar' 16:00:48 #topic Roll call 16:06:43 o7 - hello, sorry for delay priteau 16:07:23 howdy 16:07:48 So I wanted to discuss a couple of things today 16:07:54 1) Wallaby release 16:07:58 2) Xena PTG 16:08:06 and 3) AOB if there is anything else 16:08:19 #topic Wallaby release 16:09:21 The Wallaby release is coming up. Technically this is feature freeze now, but I would like to merge some of your patches so need one last push. 16:10:01 OK - which patches do you think we should focus on? 16:10:02 I've been working on the usage enforcement patch. I got unit tests fixed on the last version and I am going to test it on a real system. 16:10:45 Usage enforcement would be the main one 16:11:08 Some of the smaller ones of yours if possible 16:11:14 e.g. https://review.opendev.org/c/openstack/blazar/+/762006 16:11:16 https://review.opendev.org/c/openstack/blazar/+/766526 16:11:25 I've left a comment on that first one 16:12:30 If you get the chance of updating the soft delete patch we could look at that as well 16:13:59 Network reservation needs more work, difficult to get it merged without someone with time allocated to work on it 16:14:36 Do you think you could look at the unit test failures in https://review.opendev.org/c/openstack/blazar/+/766526? 16:15:33 yeah, re: network reservation, that's not a priority for me. i am thinking that ultimately what we really need to tackle in blazar is some actual plugin architecture. we've got 2 new resource types that i know of, and each one is so annoying to add 16:15:44 not to mention negotiate upstream patches for 16:15:50 and i don't even think they're that valuable upstream 16:16:15 i will fix up tests for CH-823002 16:16:18 grr 16:16:22 https://review.opendev.org/c/openstack/blazar/+/766526 even 16:16:51 Thanks. I've rebased the change so in one hour there should be updated results. 16:17:13 Of course you can run `tox -e py3` locally 16:19:05 I'll request a review from you when I upload the updated usage enforcement patch 16:19:10 ok 16:20:30 Now looking further ahead 16:20:34 #topic Xena PTG 16:21:11 In case you're not aware, the Project Teams Gathering for the Xena cycle will be held in April, 19-23 16:21:58 I think it would be good to discuss using the video conference tools offered by the PTG 16:22:10 It so happens that April 22 at 1600 we normally have an IRC meeting 16:22:44 that sounds like a good idea 16:23:34 There are well defined meeting slots, the best one for us is within the 13-17 UTC range 16:23:57 We could do just 16-17, but maybe booking 15-17 gives us more options? 16:24:30 agreed, 15-17, we can decide on how much of that we will use later 16:25:28 That would be 10 AM - 12 PM in Chicago if I can trust this online timezone planner 16:25:36 We will be in DST by then 16:25:41 very do-able 16:26:18 OK so I booked the Bexar (virtual Zoom) room for 15-17 UTC on April 22 16:26:35 great :) 16:26:41 I will start a document to list agenda topics 16:27:17 #action priteau Create etherpad to list Xena PTG topics 16:27:39 And I'll share the link next time 16:28:02 Feel free to invite your coworkers if they want to learn more about Blazar :) 16:28:11 :+1: 16:28:43 #topic AOB 16:29:01 Before I forget: you need to register for the PTG 16:29:11 #link https://april2021-ptg.eventbrite.com/ 16:29:16 It's free 16:29:18 i will do that today, thank you 16:29:35 will also pass this on to the rest of the team 16:29:41 It takes just a couple of minutes 16:30:19 I also had something else to share 16:31:06 I've mentioned recently, maybe just privately to you, that I saw issues with Blazar requests hanging in DevStack after the service had been up for a while. 16:31:10 A restart would fix it. 16:31:34 I thought it was due to https://review.opendev.org/c/openstack/blazar/+/771110 which changed some RPC stuff 16:31:41 But a revert didn't fix it 16:32:16 It turns out it was originating from oslo.messaging 16:32:19 https://review.opendev.org/q/Id5cddbefbe24ef100f1cc522f44430df77d217cb 16:33:05 There is a fix in a released package for master, but last time I looked it wasn't released for stable branches 16:33:28 Not sure if this is something that would affect your deployment. 16:33:50 i don't think i've seen it. is it only a thing if "direct_mandatory_flag" is set to True? 16:34:17 That's what I understand, but direct_mandatory_flag = True is the default 16:34:32 And it's not set to False by kolla-ansible 16:34:47 aha 16:34:56 So not sure why it wasn't seen on Chameleon. Maybe only affects new versions 16:35:20 Anyway, the oslo.messaging commit message hints that we're using their interface wrongly, so I will need to look into this. 16:35:26 interesting. could be. I don't think we've changed anything else in the RPC layer 16:36:00 But if you see MessageUndeliverable exceptions in logs, it could be the issue 16:36:21 I don't think I had anything else to discuss today. 16:37:25 interesting, good find 16:37:58 I don't have any other business either. maybe in the future we can talk more about the preemptible feature. it's currently being evaluated 16:38:24 I am working on this as well, but not yet ready to go upstream 16:38:54 I could share some early patches and design ideas if that's of interest to you 16:40:21 that would be good, we're actively pulling in your existing patch so any future ones would be good to know about 16:40:46 Cool. I'll make sure to share then. 16:41:52 thanks priteau! 16:41:54 Anything else? 16:42:48 nope, i think that's it 16:43:04 Thanks 16:43:15 Back to code review then :) 16:43:18 Bye 16:43:22 #endmeeting